Glioma tumor changes forecasted using MRI anchored model updates
Observation-Anchored Selective Assimilation for Longitudinal Tumor-State Proxy Forecasting in Post-Treatment Glioma
Machine LearningArtificial Intelligence
Summary
Doctors use MRI scans over time to track brain tumors after treatment, but it’s hard to predict how the tumor will change. The authors created a method that updates tumor predictions by carefully combining new MRI data with previous estimates. Their approach keeps observed tumor details as anchors and selectively adjusts other areas, which helps maintain prediction accuracy compared to simpler methods. This technique may support better monitoring of tumor changes over time using MRI images.
What this means in practice
- •For medical imaging teams: Improve longitudinal tracking of brain tumor changes by integrating intermediate MRI scans with model forecasts to enhance prediction accuracy.
- •For clinical software developers: Incorporate selective assimilation techniques into MRI analysis tools to support radiologists in monitoring glioma progression post-treatment.$Commercial implications: The method can be embedded in clinical imaging software to provide enhanced tumor forecasting features for hospitals and diagnostic centers.
Authors
Yeonjae Jung, Minwoo Shin
Abstract
Post-treatment MRI in patients with glioma provides serial observations for updating patient-specific tumor-state proxy estimates, but variable appearances and trajectories complicate forecasting. We formulate forecasting as an observation-aware digital-twin update in which an intermediate observation anchors the patient-specific state. Among 203 patients and 594 follow-up time points, a predefined no-new-treatment criterion retained 120 of 236 candidate triplets, split into 81/24/15 training/validation/test triplets at the patient level. Each time point was represented by a continuous voxel-wise tumor-state proxy map in [0,1] derived from MRI lesion labels. A SegMamba-based single-step forecaster predicted update proposals from multimodal source-state tensors. Observation-Anchored Selective Assimilation (OASA) retained the observed intermediate proxy as the state anchor and selectively applied updates through a validation-selected tiered case-level rule and voxel-wise soft gate. We compared initial-scan forecasting, rollout without assimilation, latest-observation persistence, direct prediction, OASA, OASA + calibration, and morphological dilation. Checkpoints, OASA rules, and calibration thresholds were selected using validation data only. Across three seeds on 15 held-out test triplets, OASA maintained Dice at $τ$ = 0.2 comparable to persistence (0.6071 $\pm$ 0.0025 vs. 0.6070) while yielding numerically higher Dice at $τ$ = 0.5 (0.4269 $\pm$ 0.0079 vs. 0.3981), with a small RMSE increase. Calibration increased Dice at $τ$ = 0.2 to 0.6178 $\pm$ 0.0025, increased false-positive (FP) support (11,836$\rightarrow$18,663), and reduced false-negative (FN) support (22,107$\rightarrow$17,536). This reflects near-threshold support calibration rather than improved biological predictive capability. Code is publicly available at https://github.com/jsudg436/longitudinal-proxy-forecasting.